notebooks/Spewtify.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Found 240831 playlists\n",
      "\n",
      "Already done 228907/240831\n",
      "Starting with playlist 228907\n",
      "playlist 228987/240831: 3864892 tracksretrying ...1secs\n",
      "playlist 240718/240831: 3980274 tracksMissing playlist Y by yagmuragkaya\n",
      "Missing playlist Musica para Tiendas de Ropa, comercios, etc. by luki_teverga\n",
      "Missing playlist Pop Happys  en Ingles y Latino  del 90 a la actualidad by candysumer\n",
      "playlist 240831/240831: 3981436 tracks\n",
      "TOTAL : 240831 playlists 3981436 tracks\n"
     ]
    }
   ],
   "source": [
    "# filter out Spotify (Spotify Originals?) and playlists > some size (50?)\n",
    "\n",
    "import sys\n",
    "import csv\n",
    "import spotipy\n",
    "import spotipy.util as util\n",
    "import os.path\n",
    "\n",
    "scope = 'user-library-read'\n",
    "# You can get credentials from https://developer.spotify.com/dashboard/applications\n",
    "username = 'fill this in with your details'\n",
    "client_id='fill this in with your details'\n",
    "client_secret='fill this in with your details'\n",
    "redirect_uri='fill this in with your details'\n",
    "\n",
    "csv.field_size_limit(1000000) # for really long playlists!\n",
    "\n",
    "def get_tracks_from_playlist(sp, playlist_id):\n",
    "    playlist_tracks = []\n",
    "    playlist_tracks_ids = []\n",
    "    if sp is None:\n",
    "        token = util.prompt_for_user_token(username,scope, client_id, client_secret, redirect_uri)\n",
    "        sp = spotipy.Spotify(token)\n",
    "    try:\n",
    "        results = sp.user_playlist(username, playlist_id, fields='tracks,next')\n",
    "    except spotipy.client.SpotifyException:\n",
    "        token = util.prompt_for_user_token(username,scope, client_id, client_secret, redirect_uri)\n",
    "        sp = spotipy.Spotify(token)\n",
    "        results = sp.user_playlist(username, playlist_id, fields='tracks,next')        \n",
    "    tracks = results['tracks']\n",
    "    while tracks:\n",
    "        track_ids = []\n",
    "        for item in tracks['items']:\n",
    "            track = item['track']\n",
    "            if track and 'id' in track and track['id']:\n",
    "                track_ids.append(track['id'])\n",
    "        for item in tracks['items']:\n",
    "            track = item['track']\n",
    "            if track and 'id' in track and track['id']:\n",
    "                playlist_tracks_ids.append(track['id'])\n",
    "                artist = track['artists'][0]['name']\n",
    "                if artist: artist = artist.replace(';', '')\n",
    "                else: artist = ''\n",
    "                name = track['name']\n",
    "                if name: name = name.replace(';', '')\n",
    "                else: name = ''\n",
    "                playlist_tracks.append((artist, name, track['preview_url']))\n",
    "        if tracks['next']:\n",
    "            try:\n",
    "                tracks = sp.next(tracks)\n",
    "            except:\n",
    "                token = util.prompt_for_user_token(username,scope, client_id, client_secret, redirect_uri)\n",
    "                sp = spotipy.Spotify(token)\n",
    "                tracks = sp.next(tracks)\n",
    "        else:\n",
    "            tracks = None\n",
    "    return sp, playlist_tracks_ids, playlist_tracks\n",
    "\n",
    "def get_playlists():\n",
    "    playlists_ = {}\n",
    "    # read temporary file (if exists)\n",
    "    if os.path.exists('playlists_temp.csv'):\n",
    "        with open('playlists_temp.csv', 'r',  ) as csvfile:\n",
    "            spamreader = csv.reader(csvfile, delimiter=';')\n",
    "            for row in spamreader:\n",
    "                columns = str(row)[2:-2].split(';')\n",
    "                playlists_[columns[0]] = (columns[1], columns[2], None)\n",
    "            print('Found', len(playlists_), 'playlists')\n",
    "    token = util.prompt_for_user_token(username,scope, client_id, client_secret, redirect_uri)\n",
    "    sp = spotipy.Spotify(token)\n",
    "    # output each playlist (temporarily) as name, owner\n",
    "    with open('playlists_temp.csv', 'a+', newline='', encoding='utf-8') as csvfile:\n",
    "        spamwriter = csv.writer(csvfile, delimiter=';')\n",
    "        for search in range(ord('A'), ord('z')): #A to z\n",
    "            i = 0\n",
    "            try:\n",
    "                playlists = sp.search(chr(search), 50, 0, 'playlist')\n",
    "            except spotipy.client.SpotifyException:\n",
    "                token = util.prompt_for_user_token(username,scope, client_id, client_secret, redirect_uri)\n",
    "                sp = spotipy.Spotify(token)\n",
    "                playlists = sp.search(chr(search), 50, 0, 'playlist')\n",
    "            while playlists:\n",
    "                for playlist in playlists['playlists']['items']:\n",
    "                    if not(playlist['id'] in playlists_):\n",
    "                        name = playlist['name']\n",
    "                        if name: name = name.replace(';', '')\n",
    "                        else: name = ''\n",
    "                        owner = playlist['owner']['display_name']\n",
    "                        if owner: owner = owner.replace(';', '')\n",
    "                        else: owner = ''\n",
    "                        playlists_[playlist['id']] = (name, owner, None)\n",
    "                        spamwriter.writerow([playlist['id'] + ';' + name + ';' + owner])\n",
    "                i = i + 1\n",
    "                sys.stdout.write('\\r')\n",
    "                sys.stdout.write('%c (%d/200): %d playlists' % (chr(search), i, len(playlists_)))\n",
    "                sys.stdout.flush()\n",
    "                if i < 200 and 'next' in playlists['playlists']: # i < 200\n",
    "                    try:\n",
    "                        playlists = sp.next(playlists['playlists'])\n",
    "                    except spotipy.client.SpotifyException:\n",
    "                        token = util.prompt_for_user_token(username,scope, client_id, client_secret, redirect_uri)\n",
    "                        sp = spotipy.Spotify(token)\n",
    "                        playlists = sp.next(playlists['playlists'])\n",
    "                else:\n",
    "                    playlists = None\n",
    "    print()\n",
    "    return playlists_\n",
    "\n",
    "def get_tracks_from_playlists(playlists):\n",
    "    sp = None\n",
    "    # read file (if exists)\n",
    "        tracks_ = {}\n",
    "        if os.path.exists('tracks.csv'):\n",
    "            with open('tracks.csv', \"r\", encoding='utf-8') as csvfile:\n",
    "                spamreader = csv.reader(csvfile, delimiter=';')\n",
    "                for row in spamreader:\n",
    "                    columns = str(row)[2:-2].split(';')\n",
    "                    tracks_[columns[0]] = (columns[1], columns[2], columns[3])\n",
    "    already_done = 0\n",
    "    done_ids = {}\n",
    "    if os.path.exists('playlists.csv'):\n",
    "        with open('playlists.csv', \"r\", encoding='utf-8') as csvfile:\n",
    "            spamreader = csv.reader(csvfile, delimiter=';')\n",
    "            for row in spamreader:\n",
    "                columns = str(row)[2:-2].split(';')\n",
    "                done_ids[columns[0]] = 1\n",
    "                already_done = already_done + 1\n",
    "    print('Already done %d/%d' % (already_done, len(playlists)))\n",
    "    # output each track as id, artist name, name, preview url\n",
    "    with open('tracks.csv', 'a+', newline='', encoding='utf-8') as csvfile_tracks:\n",
    "        spamwriter_tracks = csv.writer(csvfile_tracks, delimiter=';')\n",
    "        with open('playlists.csv', 'a+', newline='', encoding='utf-8') as csvfile_playlists:\n",
    "            spamwriter_playlists = csv.writer(csvfile_playlists, delimiter=';')\n",
    "            caught_up = 0\n",
    "            for j, key in enumerate(playlists.keys()):\n",
    "                if not(caught_up):\n",
    "                    if j < already_done or key in done_ids: # might have had to skip some\n",
    "                        continue\n",
    "                    else:\n",
    "                        print('Starting with playlist', j)\n",
    "                        caught_up = 1\n",
    "                try:\n",
    "                    sp, playlist_tracks_ids, playlist_tracks = get_tracks_from_playlist(sp, key)\n",
    "                except spotipy.client.SpotifyException:\n",
    "                    print('Missing playlist', playlists[key][0], 'by', playlists[key][1])\n",
    "                    continue\n",
    "                for i in range(0, len(playlist_tracks_ids)):\n",
    "                    track_key = playlist_tracks_ids[i]\n",
    "                    if not(track_key in tracks_):\n",
    "                        tracks_[track_key] = playlist_tracks[i]\n",
    "                        if tracks_[track_key][2]:\n",
    "                            url = tracks_[track_key][2]\n",
    "                        else:\n",
    "                            url = ''\n",
    "                        spamwriter_tracks.writerow([track_key + ';' + tracks_[track_key][0] + ';' + tracks_[track_key][1] + ';' + url + ';'])\n",
    "                playlists[key] = (playlists[key][0], playlists[key][1], playlist_tracks_ids)\n",
    "                playlist = playlists[key]\n",
    "                if (len(playlist[2]) > 0):\n",
    "                    tracks = ''\n",
    "                    for i in range(len(playlist[2]) - 1):\n",
    "                        tracks = tracks + str(playlist[2][i]) + ';'\n",
    "                    tracks = tracks + str(playlist[2][len(playlist[2]) - 1])\n",
    "                    spamwriter_playlists.writerow([playlist[0] + ';' + playlist[1] + ';' + tracks])\n",
    "                sys.stdout.write('\\r')\n",
    "                sys.stdout.write('playlist %d/%d: %d tracks' % (j + 1, len(playlists), len(tracks_)))\n",
    "                sys.stdout.flush()\n",
    "    print()\n",
    "    return tracks_\n",
    "\n",
    "# main\n",
    "playlists = get_playlists()\n",
    "tracks = get_tracks_from_playlists(playlists)\n",
    "print('TOTAL :', len(playlists), 'playlists', len(tracks), 'tracks')"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.6.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}